INSTITUTO POTOSINO DE INVESTIGACIÓN
CIENTÍFICA Y TECNOLÓGICA, A.C.
POSGRADO EN CIENCIAS APLICADAS
Carbon nanostructured adsorbents for the removal of
toxic metals from aqueous solution
Tesis que presenta
Nancy Verónica Pérez Aguilar
Para obtener el grado de
Doctor en Ciencias Aplicadas
En la opción de
Ciencias Ambientales
Director de la Tesis:
José René Rangel Méndez
ii
iii
CRÉDITOS INSTITUCIONALES
Esta tesis fue elaborada en la División de Ciencias Ambientales, con apoyo del Laboratorio de Investigación en Nanociencias y Nanotecnología (LINAN), del Instituto Potosino de Investigación Científica y Tecnológica, A.C., bajo la dirección del Dr. José René Rangel Méndez.
Durante la realización del trabajo el autor recibió una beca académica del Consejo Nacional de Ciencia y Tecnología (CONACYT No. 204214)
Este trabajo de investigación fue financiado por: Fondos Mixtos CONACYT-Estado de Puebla (PUE-2004-C02-5) y Fondos CONACYT-Investigación Básica (SEP-2004-C01-45764).
v
DEDICATORIA
A la memoria de papá, Domitilo Pérez García (q.e.p.d.)
Papá por treinta o cuarenta años, amigo de mi vida todo el tiempo protector de mi miedo, brazo mío,
palabra clara, corazón resuelto.
Algo le falta al mundo, y tú te has puesto a empobrecerlo mas, y a hacer a solas
tus gentes tristes y tu Dios contento.
Y es en vano llorar. Y si golpeas las paredes de Dios, y si te arrancas
el pelo o la camisa,
nadie te oye jamás, nadie te mira. No vuelve nadie, nada. No retorna
el polvo de oro a la vida.
Jaime Sabines
Con amor a mi familia:
a mamá Ma. de Jesús Aguilar a Gaby, Pepe, Lilí, Tito
vi
AGRADECIMIENTOS
Al Dr. René Rangel Méndez por la oportunidad de desarrollar este proyecto. Al Dr. Vladimir Escobar por su valiosa asesoría en el área de los polímeros.
A los doctores Emilio Muñoz de la División de Materiales Avanzados del IPICYT, y Roberto Leyva Ramos de la Facultad de Ciencias Químicas de la UASLP, por su participación en el comité tutoral y como sinodales de la presente tesis.
A Sydney Robertson-Jiménez por su valioso apoyo técnico en el idioma Inglés. Al Laboratorio de Investigación en Nanociencias y Nanotecnología (LINAN) del Instituto Potosino de Investigación Científica y Tecnológica, A.C. por las facilidades proporcionadas para el acceso y utilización de los diferentes equipos. Al personal técnico de los diferentes laboratorios del IPICYT por su valiosa asistencia técnica: Daniel Ramírez, Grisel Ramírez, Hugo Martínez, Sofía Vega, Ana Laura Elías, Jessica Campos, Magdalena Martínez, Rebeca Pérez, Griselda Chávez, Dulce Partida y Rosy Martínez.
A los doctores Marco Martin González y Selene Berber de la Facultad de Ciencias Químicas de la UASLP, al Dr. Rafael Herrera, Dr. Rodolfo Ruiz y Daniuska Escobar de la Facultad de Química de la UNAM, al Dr. Julio Soto de 3M de México, por las facilidades proporcionadas para utilizar sus instalaciones y equipos.
A mis amigas y amigos por los inolvidables momentos compartidos.
vii
TABLE OF CONTENTS
CONSTANCIA DE APROBACIÓN DE LA TESIS ii
CRÉDITOS INSTITUCIONALES iii
ACTA DE EXAMEN iv
DEDICATORIAS v
AGRADECIMIENTOS vi
LIST OF TABLES xi
LIST OF FIGURES xiii
SUPPORTING INFORMATION xvii
ABSTRACT xx
RESUMEN xxii
NOMENCLATURE xxiv
CHAPTER 1
Carbon Nanostructured Adsorbents for the Removal of Toxic Metals from Aquoeus Solution. State of the Art
1
1.1 INTRODUCTION 1
1.1.1 Water Pollution by Toxic Metals and Technologies for Treatment 2
1.1.1.1 Water Pollution by Lead 2
1.1.1.2 Water Pollution by Cadmium 4
1.1.1.3 Water Treatment Technologies Available to Remove
Toxic Metals 6
1.1.1.4 Adsorbents for Toxic Metals 9
1.1.1.5 Adsorption of Toxic Metals on Activated Carbon 10 1.2 NANOTECHNOLOGY DEVELOPMENT AND POTENTIAL
APPLICATIONS TO WATER TREATMENT 13
viii
1.3 MOTIVATION FOR THIS RESEARCH 23
1.4 GENERAL AND SPECIFIC OBJECTIVES 24
1.5 HYPOTHESIS 25
1.6 STRUCTURE OF THE THESIS 25
CHAPTER 2
Adsorption of Cadmium and Lead onto Oxidized Nitrogen-doped Multiwall Carbon Nanotubes in Aqueous Solution: Equilibrium and Kinetic
27
Abstract 27
2.1 INTRODUCTION 28
2.2 EXPERIMENTAL 30
2.2.1 Synthesis and Chemical Oxidation of CNx 30
2.2.2 Characterization of Pristine and Oxidized CNx 30
2.2.2.1 Morphological Characterization 30
2.2.2.2 Physical and Chemical Characterization 31 2.2.3 Adsorption/Desorption and Selectivity of Cadmium and Lead 32
2.2.4 Kinetic Experiments 33
2.3 RESULTS AND DISCUSSION 34
2.3.1 Synthesis of Nitrogen-doped Carbon Nanotubes (CNx) and their Oxidation by Nitric Acid
34
2.3.2 Physical and Chemical Properties of Pristine and Oxidized Carbon Nanotubes
34
2.3.3 Surface Chemistry of CNx Nanotubes and their Cadmium and Lead Adsorption Capacity
43
2.3.4 Cadmium Adsorption Kinetic onto Oxidized CNx 48
2.4 CONCLUSIONS 50
CHAPTER 3
Morphology Effect of Three Different Types of Carbon Nanotubes on Cadmium Adsorption Kinetic
ix
Abstract 51
3.1 INTRODUCTION 52
3.2 EXPERIMENTAL 55 3.2.1 Materials and Oxidation of Carbon Nanotubes 55
3.2.2 Characterization of Materials 55
3.2.3 Adsorption Equilibrium Experiments 56
3.2.4 Adsorption Kinetic Experiments 56
3.2.5 Modeling of Adsorption Kinetic Data 58
3.2.5.1 Pseudo-second Order Kinetic Model 59
3.2.5.2 External Mass Transfer Model 60
3.2.5.3 Intraparticle Diffusion Model 62
3.3 RESULTS AND DISCUSSION 63
3.3.1 Chemical Modification and Physicochemical Characterization of Carbon Nanotubes and Iron Oxide Nanoparticles
63
3.3.2 Cadmium Adsorption 74
3.3.3 Cadmium Adsorption Kinetic onto Oxidized Carbon Nanotubes 76 3.3.3.1 Analysis of Adsorption Kinetic Data with the
Pseudo-second Order Model
79
3.3.3.2 Analysis of Adsorption Kinetic Data with an External
Mass Transfer Model 83
3.3.3.3 Analysis of Adsorption Kinetic Data with an Intraparticle
Diffusion Model 87
3.4 CONCLUSIONS 93 CHAPTER 4
Carbon Nanotube-Based Composites for Adsorption of Toxic Metals from Aqueous Solution
95
Abstract 95
4.1 INTRODUCTION 96
x
4.2.1 Materials and Synthesis of Carbon Nanotube-Based
Composites 98
4.2.2 Characterization of Carbon Nanotube-Based Composites 99
4.2.3 Adsorption Equilibrium Experiments 100
4.3 RESULTS AND DISCUSSION 101
4.3.1 Synthesis and Characterization of Carbon Nanotube-Based Composites
101
4.3.2 Adsorption Behavior of Carbon Nanotube-Based Composites 111 4.3.2.1 Cadmium Adsorption Capacity of Carbon
Nanotube-Based Composites
111
4.3.2.2 Cadmium Adsorption Tests in Packed Column with Carbon Nanotube-Based Composites
113
4.4 CONCLUSIONS 118 CHAPTER 5
General Discussion, Final Conclusions and Perspectives for Future
Research 120
5.1 GENERAL DISCUSSION 120
5.2 FINAL CONCLUSIONS 125
5.3 PERSPECTIVES FOR FUTURE RESEARCH 128
REFERENCES 130
LIST OF PUBLICATIONS 147
EXTENDED ABSTRACTS 147
ATTENDANCE AT CONFERENCES 148
xi
LIST OF TABLES
Table 2.1 Surface area and pore volume of micropores (dp < 2nm) and mesopores (2 nm < dp < 50 nm) of pristine CNx and oxidized CNx by periods of 1 h, 3 h, and 5 h, determined by nitrogen adsorption at 77 K
37
Table 2.2 Chemical composition of pristine CNx and nitric acid oxidized CNx at 80 ± 3 °C
39 Table 2.3 Concentration of total acid sites (TAS) of pristine nanotubes
and oxidized CNx for 1 h, 3 h and 5 h, determined by Boehm’s titration method at 25 °C
40
Table 3.1 Spacing between graphene layers in carbon nanotubes determined by X-ray diffraction
66 Table 3.2 Chemical composition of nitric acid oxidized carbon nanotubes 71 Table 3.3 Langmuir isotherm parameters determined for cadmium
adsorption onto oxidized carbon nanotubes and iron oxide nanoparticles, at pH 6 and 25 °C
75
Table 3.4 Experimental conditions for batch adsorption kinetic tests 77 Table 3.5 Pseudo-second order parameters for adsorption kinetic of
cadmium onto oxidized carbon nanotubes and iron oxide nanoparticles
80
Table 3.6 Physical properties of single nanoparticles to predict concentration decay curves by using the external mass transfer model
84
Table 3.7 External mass-transfer parameters for adsorption kinetic of cadmium onto oxidized carbon nanotubes and iron oxide nanoparticles
84
Table 3.8 Physical properties of nanoparticles considered for the intraparticle diffusion model
87 Table 3.9 Intraparticle diffusion model parameters for adsorption kinetic
of cadmium onto single oxidized carbon nanotubes and iron oxide nanoparticles
88
Table 3.10 Intraparticle diffusion model parameters for adsorption kinetic of cadmium onto agglomerates of carbon nanotubes and iron oxide nanoparticles
88
Table 4.1 BET surface area and density of oxidized nanotubes, polyurethane matrix and obtained nanotube-based composites, determined by nitrogen adsorption at 77 K, and by helium displacement, respectively
xii
Table 4.2 Glass transition temperature (Tg) and melting temperature (Tm) of the obtained composites determined by differential scanning calorimetry (DSC)
110
Table 4.3 Langmuir isotherm parameters determined for cadmium adsorption onto nanotube-based composites, at pH 6 and 25 °C
112
Table 4.4 Experimental conditions for dynamic adsorption tests in bed-packed columns with composites C1 and C3, for cadmium adsorption in aqueous solution
114
Table 4.5 Dynamic adsorption tests for composites C3 and C1 until bed saturation. Influent cadmium solution at pH 6 and room temperature (22 ± 3 °C)
xiii
LIST OF FIGURES
Fig. 1.1 Species distribution of 4x10-4 M Pb (II) in aqueous solution at 25 °C.
4 Fig. 1.2 Species distribution of 7x10-4 M Cd (II) in aqueous solution at 25
°C.
5 Fig. 1.3 Representation of oxygen functional groups present on the
carbon activated surface.
11 Fig. 1.4 Fundamental structure for (A) graphene sheet, (B) single-wall
carbon nanotube, SWNT, (C ) multiwall carbon nanotube, MWNT (modified figure from Mauter and Elimelech, 2008) and (D) typical defects and functional groups in an oxidized carbon nanotube (modified figure from Hirsch, 2002).
18
Fig. 2.1 a,b SEM images of pristine CNx. c STEM image of a tip of CNx before oxidation with nitric acid. d, e SEM images of nitric acid oxidized CNx by 5 h at 80 °C. f STEM image of the tips of nitric acid oxidized CNx by 5 h at 80 °C. The circles at the tips of the nanotubes show some changes due to the oxidative process.
35
Fig. 2.2 XRD spectra for pristine CNx and nitric acid oxidized CNx by 5 h (ox-CNx-5h), at 80 ± 3°C.
36 Fig. 2.3 RAMAN spectra of pristine nanotubes and nitric acid oxidized
CNx for 1 h, 3 h, and 5 h at 80 ± 3 °C.
38 Fig. 2.4 FTIR spectra of (a) pristine CNx and (b) nitric acid oxidized CNx
by 5 h at 80 ± 3°C.
41 Fig. 2.5 Mass-loss profiles for (a) pristine nanotubes and (b) oxidized
CNx by 5 h, heated up from 40 °C to 1000 °C under nitrogen atmosphere. Decomposition rates derived from mass-loss profiles of the same samples were included and labeled as (c) and (d), respectively.
42
Fig. 2.6 Proton-binding curves for pristine nitrogen-doped carbon nanotubes (CNx) and oxidized nanotubes (ox-CNx) determined at 25 °C and 0.05 M ionic strength.
43
Fig. 2.7 Effect of pH on the adsorption isotherms of (a) lead and (b) cadmium onto oxidized CNx determined at 25 °C.
45 Fig. 2.8 Competition between adsorption of (a) lead and (b) cadmium
onto oxidized CNx at pH 5 and 25 °C. Individual isotherms for (c) lead and (d) cadmium at pH 5 and 25° C were included.
46
Fig. 2.9 Adsorption isotherms of cadmium in solution at pH 6 and 25 °C, on (a) ox-CNSW, (b) ox-CNx and (c) ox-ACF.
xiv
Fig. 2.10 Kinetic experiments with oxidized CNx (circles) and powder activated carbon (PAC-F400, squares) for cadmium in solution at pH 6 and room temperature.
49
Fig. 3.1 Configuration of stirred vessel for adsorption kinetic experiments 57 Fig. 3.2 Mass transport processes on cadmium adsorption by carbon
nanotubes. (I) Adsorption on a carbon nanotube includes: (1) external mass transport, (2) pore volume diffusion through axial and/or radial direction (3) adsorption, and (4) surface diffusion. (II) Adsorption on aggregates of carbon nanotubes, where pore volume diffusion also includes the space between nanotubes.
58
Fig. 3.3 SEM images of used nanoparticles in adsorption kinetic experiments. (a) Bundles of pristine aligned multiwall carbon nanotubes, (b) entangled ropes of single-wall carbon nanotubes; (c) exfoliated bundles of oxidized nitrogen-doped carbon nanotubes; (d) catalyst free and eroded tip of an oxidized nitrogen-doped carbon nanotube; (e) partial destruction of an oxidized multiwall carbon nanotube; (f) agglomerate of iron oxide nanoparticles.
65
Fig. 3.4 XRD patterns show the interlayer spacing between graphene layers for oxidized carbon nanotubes: (a) single-wall nanotubes, ox-SWNT, (b) nitrogen-doped nanotubes, ox-CNx, and (c) multiwall nanotubes, ox-MWNT.
66
Fig. 3.5 XRD patterns show shifting of the plane (002) corresponding to the interlayer spacing between graphene layers, for (a) pristine doped carbon nanotubes, CNx, (b) oxidized nitrogen-doped carbon nanotubes, ox-CNx, and (c) cadmium exhausted ox-CNx.
67
Fig. 3.6 RAMAN spectra obtained with 514 nm laser line, for oxidized carbon nanotubes: (a) single-wall nanotubes, ox-SWNT, (b) nitrogen-doped nanotubes, ox-CNx and (c) multiwall nanotubes, ox-MWNT.
68
Fig. 3.7 Pore size distribution of (a) SWNT; (b) CNx; (c) ox-MWNT; and (d) iron oxide nanoparticles.
70 Fig. 3.8 Oxygen acidic groups reported as total acidic sites (TAS),
carboxylic, phenolic and lactonic groups attached onto oxidized carbon nanotubes, determined by titration.
72
Fig. 3.9 FTIR spectra of (a) oxidized nitrogen-doped carbon nanotubes, ox-CNx; (b) oxidized multiwall carbon nanotubes, ox-MWNT; (c) oxidized single-wall carbon nanotubes, ox-SWNT; and (d) iron oxide nanoparticles, FeOOH, obtained by attenuated total reflectance (ATR).
xv
Fig. 3.10 (a) HRTEM image of oxidized CNx reveal a structure of cadmium, the inset shows its interlayer spacing (about 0.26 nm); (b) STEM image shows a surface mapping analysis for ox-CNx to corroborate the presence of cadmium (showed in purple), carbon is shown in red.
74
Fig. 3.11 Cadmium adsorption isotherms at pH 6 and 25 °C adjusted by Langmuir model. (a) Oxidized single-wall carbon nanotubes, ox-SWNT; (b) oxidized nitrogen-doped carbon nanotubes, ox-CNx; (c) oxidized multiwall carbon nanotubes, ox-MWNT; (d) iron oxide nanoparticles, FeOOH
76
Fig. 3.12 HRTEM Micrographs of tips and internal tubular section of (a) oxidized single-wall carbon nanotubes, ox-SWNT; (b) oxidized multiwall carbon nanotubes, ox-MWNT; and (c) oxidized nitrogen-doped carbon nanotubes, ox-CNx.
78
Fig. 3.13 Cadmium adsorption rate onto (a) SWNT, run 13; (b) ox-CNx, run 6; (c) ox-MWNT, run 10; and (d) FeOOH, run 16, fitted by the pseudo-second order model. Data obtained at CA0 4 mg/L, m/V=0.4 g/L, pH 6 ± 0.1, 25 °C, 200 rpm. Inlet shows cadmium uptake curves predicted with model parameters.
81
Fig. 3.14 Curves for approximation to equilibrium (lines) predicted by the pseudo-second order model, showing the half-life adsorption time t0.5 (empty symbols), and the time to attain 0.95 of
adsorption t0.95 (full symbols), for cadmium adsorption onto (a)
ox-SWNT (run 13); (b) ox-CNx (run 6); (c) ox-MWNT (run 10); and (d) FeOOH (run 16).
82
Fig. 3.15 Concentration decay curves for cadmium adsorption onto (a) ox-SWNT, run 13; (b) ox-CNx, run 6; (c) ox-MWNT, run 10; and (d) FeOOH, run 16, predicted by the external mass transfer model at CA0 4 mg/L, pH 6 ± 0.1, 25 °C and 200 rpm. Symbols and lines represent the experimental data and the predicted curves respectively.
85
Fig. 3.16 Concentration decay curves predicted by the intraparticle diffusion model, for cadmium adsorption onto (a) ox-SWNT, run 13; (b) ox-CNx, run 6; (c) ox-MWNT, run 10; and (d) FeOOH, run 16. Symbols represent experimental data and the lines the model considering single or agglomerates of nanoparticles.
90
Fig. 4.1 SEM images show (a) oxidized nitrogen-doped carbon nanotubes (ox-CNx) before being supported; surface fracture of (b) polyurethane foam matrix; (c) the composite C1 with nanotube load of 2.5%; (d) holes caused by the extraction of carbon nanotubes (dotted circles) in the composite C1; (e) the composite C2 with nanotube load of 5%, and outer surface of the composite (inset); (f) segregated sections of SBR (dotted
xvi
circles) and carbon nanotubes (arrows) in the composite C3 with nanotube loading of 1% plus 5% SBR.
Fig. 4.2 XRD spectra for nitric acid oxidized carbon nanotubes, ox-CNx, polyurethane matrix (PU and additive containing PU-SBR), and obtained composites with nanotube load of 2.5% (C1), 5% (C2), and 1%, C3.
105
Fig. 4.3 FTIR spectra obtained by attenuated total reflectance (ATR) for (a) oxidized carbon nanotubes (ox-CNx) and polyurethane matrix (PU), (b) PU and obtained composites with nanotube load of 2.5% (C1) and 5% (C2), and (c) PU-SBR matrix and obtained composite with nanotube load of 1% (C3); C3-Cd is the spectra for cadmium adsorbed onto composite C3.
106
Fig. 4.4 Raman spectra at 632.8 nm excitation wavelength of oxidized nanotubes (ox-CNx), polyurethane matrix (PU) and the obtained composites with nanotube load of 2.5% (C1) and 5% (C2).
107
Fig. 4.5 Thermograms for (a) polyurethane matrix, PU and composites C1 and C2 with nanotube load of 2.5% and 5%, respectively; and (b) PU-SBR matrix, oxidized nanotubes (ox-CNx) and composite with nanotube load of 1% (C3). Insets show DTA analyses for each sample.
109
Fig. 4.6 Cadmium adsorption isotherms at pH 6 and 25 °C adjusted by Langmuir model, determined for polyurethane matrix (PU) and composites with nanotube load of 2.5% (C1), 5% (C2) and 1% (C3). The inset shows isotherms at low concentration.
112
Fig. 4.7 Cadmium breakthrough curves obtained for composite C3 at EBCT of 10 min (Run 1) and 20 min (Run 2). Initial concentration of 0.238 and 0.178 mg/L, respectively, at pH 6 and room temperature.
115
Fig. 4.8 Consecutive Cd2+ breakthrough curves and adsorption/desorption processes using nanotube-based composites packed in a column. Composites C1 and C3 with nanotubes load of 2.5% and 1% + SBR, respectively. Cadmium influent at 0.3 ± 0.1 mg/L, pH 6, room temperature and 3 BV/h.
xvii
SUPPORTING INFORMATION
Appendix A
Fig. A1 Adsorption kinetic data for cadmium adsorption onto ox-CNx without pH control, fitted to pseudo-second order model. Runs 1 and 2 were performed at CA0~ 4 mg/L and 150 rpm; Runs 3 to 5 were performed at 200 rpm and CA0 near 4, 9 and 19 mg/L respectively.
149
Fig. A2 Adsorption kinetic data for cadmium adsorption onto ox-CNx at pH 6, fitted to pseudo-second order model. Runs 6 to 8 were performed at 200 rpm and CA0 near 4, 9 and 19 mg/L respectively.
150
Fig. A3 Adsorption kinetic data for cadmium adsorption onto ox-MWNT fitted to pseudo-second order model. Cadmium initial concentration CA0 was approximately 4 mg/L; Run 9 was
performed without pH control and 200 rpm, Run 10 was performed at pH 6 and 200 rpm, Run 11 was performed at pH 6 and 250 rpm.
151
Fig. A4 Adsorption kinetic data for cadmium adsorption onto ox-SWNT fitted to pseudo-second order model. Cadmium initial concentration CA0 was approximately 4 mg/L; Run 12 was
performed without pH control and 200 rpm, Run 13 was performed at pH 6 and 200 rpm, Run 14 was performed at pH 6 and 350 rpm.
152
Fig. A5 Adsorption kinetic data for cadmium adsorption onto non-porous iron nanoparticles, FeOOH, fitted to pseudo-second order model. Cadmium initial concentration CA0 was
approximately 4 mg/L; Run 15 was performed without pH control and 200 rpm, Run 16 was performed at pH 6 and 200 rpm.
153
Appendix B
Fig. B1 Concentration decay curves with experimental data fitted by the external mass transfer model for cadmium adsorption onto ox-CNx. Runs were performed without pH control. (a) Runs 2 and 3 were performed at CA0~ 4 mg/L and 150 rpm; (b) Runs 3, 4 and 5 were performed at 200 rpm and CA0 near 4, 9 and 19 mg/L respectively.
154
Fig. B2 Concentration decay curves with experimental data fitted by the external mass transfer model for cadmium adsorption
xviii
onto ox-CNx. Runs 6, 7 and 8 were performed at initial cadmium concentration of approximately CA0 4, 9 and 19 mg/L respectively, at pH 6 and 200 rpm.
Fig. B3 Concentration decay curves with experimental data fitted by the external mass transfer model for cadmium adsorption onto ox-MWNT at initial cadmium concentration of approximately CA0 4 mg/L. Run 9 was performed without pH
control and 150 rpm; Run 10 was performed at pH 6 and 200 rpm; Run 11 was performed at pH 6 and 250 rpm.
156
Fig. B4 Concentration decay curves with experimental data fitted by the external mass transfer model for cadmium adsorption onto ox-SWNT at initial cadmium concentration of approximately CA0 4 mg/L. Run 12 was performed without pH
control and 200 rpm; Run 13 was performed at pH 6 and 200 rpm; Run 14 was performed at pH 6 and 350 rpm.
157
Fig. B5 Concentration decay curves with experimental data fitted by the external mass transfer model for cadmium adsorption onto FeOOH at initial cadmium concentration of approximately CA0 4 mg/L. Run 15 was performed without pH
control and 200 rpm; Run 16 was performed at pH 6 and 200 rpm.
158
Appendix C
Fig. C1 Sensitivity test of the intraparticle diffusion model to predict concentration decay curves (lines) that best fitting to the experimental adsorption kinetic data (symbols). Different values of diffusivity coefficients (De,p) were tested for run 1,
performed at initial cadmium concentration of 4 mg/L, initial pH 6 without control and 150 rpm.
159
Fig. C2 Experimental adsorption kinetic data (symbols) and concentration decay curves (lines) predicted by the intraparticle diffusion model, for cadmium adsorption onto ox-CNx. Runs 3, 4 and 5 were performed at 200 rpm and (a) CA0=4.14 mg/L without pH control, (b) CA0 =9.65 mg/L at pH 6
and (c) CA0 =19.29 mg/L at pH 6.
160
Fig. C3 Experimental adsorption kinetic data (symbols) and concentration decay curves (lines) predicted by the intraparticle diffusion model, for cadmium adsorption onto ox-CNx. Runs 6, 7 and 8 were performed at 200 rpm, pH 6 and (a) CA0 =4.14 mg/L, (b) CA0 =9.65 mg/L and (c) CA0 =19.29
mg/L.
xix
Fig. C4 Effect of pH control on the decay concentration curves with experimental data fitted by the intraparticle diffusion model, for cadmium adsorption onto (a) ox-SWNT, Runs 12 and 13, (b) ox-MWNT, Runs 9 and 10, and (c) ox-CNx, Runs 3 and 6. Experimental data were obtained at cadmium initial concentration of 4 mg/L, 200 rpm, without pH control (empty symbols) or pH 6 (full symbols).
162
Fig. C5 Effect of shaking rate on the decay concentration curves with experimental data fitted by the intraparticle diffusion model. Experimental data were obtained at initial cadmium concentration of 4 mg/L, for cadmium adsorption onto (a) ox-SWNT, runs 13 and 14, at 200 and 350 rpm respectively, (b) ox-MWNT, runs 10 and 11, at 200 and 250 rpm respectively, and (c) ox-CNx, runs 2 and 3, at 150 and 200 rpm respectively. Empty symbols represent low rate data and full symbols high rate data.
xx
ABSTRACT
Nowadays most bodies of water are polluted by toxic heavy metals, which persist for long periods even at low concentrations. Permissible limits for toxic heavy metals in drinking water are continuously reviewed to protect human health. Accomplishment of established regulations requires the improvement of current technologies and the development of new ones. Nanotechnology has the potential to preserve water quality enabling its reuse, through development of nanoparticles to detect, prevent, and remove toxic heavy metals by adsorption. However, nanoparticles should be immobilized prior to be applied to water treatment to prevent their liberation and negative impact on the environment.
Early adsorption studies of toxic metals (cadmium, lead, copper, nickel, zinc and chromium) were performed with oxidized carbon nanotubes of single wall (SWNT) and multiwall (MWNT). Reported results showed that compared to activated carbon, carbon nanotubes had higher adsorption capacity, shorter equilibrium time and the possibility of being regenerated and used through several cycles.
In this study, oxidized nitrogen-doped multiwall carbon nanotubes (ox-CNx) showed 1.8 and 1.4 times higher adsorption capacity for cadmium and lead than ox-MWNT and oxidized granular activated carbon (ox-ACF), but 0.65 times lower than for ox-SWNT, at ph 5 and 25 °C. The small size, geometry and surface chemical composition of ox-CNx are the key factors for their higher adsorption capacity than ox-MWNT and activated carbon. Metal-exhausted ox-CNx could be regenerated and reused since more than 90% of the mass of both metals adsorbed was desorbed.
xxi
experimental kinetic data with lower deviations than 5% and calculated diffusivity coefficients of about 2.4E-11, 4.2E-9 and 1.1E-9 cm2/s, for ox-SWNT, ox-CNx and ox-MWNT, respectively, considered as single nanotubes, at initial cadmium concentration of 4 mg/L, pH 6 and 200 rpm.
Nanoestructured adsorbents were obtained supporting oxidized carbon nanotubes (ox-CNx) in polyurethane: nanotube loading of 2.5% (composite C1) and 1% plus 5% SBR (composite C3). Dynamic adsorption tests were performed in bed-packed columns with composites C1 and C3; about 5 and 7 bed volumes of cadmium solution were processed by C1 and C3, respectively, before bed saturation. Physical and chemical interactions between oxygen surface groups of carbon nanotubes and polymer chains could disable adsorption sites of carbon nanotubes. Consecutive cycles of adsorption-desorption suggested that nanotubes were firmly supported by the polymeric matrix. Obtained composites can be used in water treatment systems while preventing carbon nanotubes dispersion. Further research is required to effectively support carbon nanotubes or other nanoparticles preserving their adsorptive features.
xxii
RESUMEN
En la actualidad gran parte de los cuerpos de agua superficiales y subterráneos están contaminados por metales tóxicos, lo que implica un riesgo para la salud humana y la biodiversidad de los ecosistemas. Para obtener la calidad del agua establecida en las normas, se requieren tecnologías de purificación y tratamiento cada vez más eficientes. La nanotecnología ha surgido como una alternativa para preservar la calidad de este recurso.
En la presente investigación se probaron nanotubos de carbono dopados con nitrógeno oxidados (ox-CNx), para adsorber cadmio y plomo en solución acuosa. La capacidad máxima de adsorción de cadmio (qm) sobre ox-CNx fue 1.8 y 1.4
veces mayor que la de ox-MWNT y carbón activado oxidado (ox-ACF), respectivamente, pero 0.65 veces menor que la de ox-SWNT. Factores como la geometría, la composición química y menor tamaño de partícula de los nanotubos ox-CNx pudieron determinar su mayor capacidad de adsorción de cadmio en relación con ox-MWNT y carbón activado. Además, es posible desorber más del 90% de los metales adsorbidos utilizando soluciones ácidas (pH 2) y reutilizar estos nanotubos (ox-CNx) durante varios ciclos, manteniendo su capacidad de adsorción.
xxiii
modelo fueron 4.2E-9, 1.1E-9 y 2.4E-11 cm2/s para nanotubos individuales de ox-CNx, ox-MWNT y ox-SWNT, respectivamente, a una concentración inicial de cadmio de 4 mg/L, a pH 6 y 200 rpm.
Los nanotubos ox-CNx se inmovilizaron en poliuretano para obtener un adsorbente nanoestructurado. Estos nanotubos se seleccionaron considerando su mayor capacidad de adsorción de cadmio que los ox-MWNT, además de su cinética de adsorción más rápida que la de los ox-MWNT y ox-SWNT. Los compositos preparados con ox-CNx tuvieron concentraciones de 2.5% y 5% en peso (compositos C1 y C2), además de 1% en peso de nanotubos más 5% de SBR (composito C3). En el composito ocurrieron interacciones físicas y químicas entre los grupos superficiales de los nanotubos y el polímero. Como resultado, la concentración de sitios de adsorción disponibles en la superficie de los nanotubos disminuyó de manera significativa. Los compositos C1 y C3 se empacaron en columnas de lecho fijo para realizar pruebas de adsorción dinámicas; C1 y C3 procesaron aproximadamente 5 y 7 volúmenes de lecho de solución de cadmio, respectivamente, antes de la saturación del lecho. Ciclos consecutivos de adsorción-desorción mostraron que los nanotubos permanecieron en la matriz polimérica. Los compositos obtenidos son un medio de soporte seguro que permitirían la aplicación de nanotubos de carbono en sistemas de tratamiento de agua, evitando su dispersión. Es necesario continuar investigando cómo incrementar el porcentaje de nanotubos de carbono en los compositos a la vez mantener sus propiedades de adsorción.
xxiv
NOMENCLATURE
AAS Atomic absorption spectroscopy
AC Activated carbon
GAC Granular activated carbon Filtrasorb® 400 from Calgon Carbon Corporation
ATRFTIR Fourier transformed infrared spectroscopy by attenuated total reflectance
BV Bed-packed volume (cm3)
C Solution concentration (mg/L)
CA Concentration of cadmium in solution at time t
(mg/cm3)
CAo Initial concentration of cadmium in solution
(mg/cm3)
CA∞ Concentration of cadmium at equilibrium (mg/cm3)
CA,r Concentration of cadmium in the pore volume at a
distance r (mg/cm3 )
CA,R Concentration of cadmium in solution at the
external surface (mg/cm3)
C0 Initial cadmium concentration (mg/L)
Ce Concentration of cadmium solution at equilibrium
(mg/L)
Ci Concentration of cadmium solution at time interval i
(mg/L)
CS Equilibrium concentration of cadmium on the
surface of solid nanoparticle (mg/cm3)
CNx Nitrogen-doped carbon nanotubes
Dcorr Dimensionless effective diffusion coefficient
(cm2/h)
xxv
Di Ionic diffusivity of cadmium (cm2/h)
D Ionic diffusivity of cadmium at diluted
concentrations (cm2/h)
Ds Intraparticle diffusion coefficient (cm2/h)
dp Nanoparticle diameter (cm)
DTA Differential thermal analysis
DRIFTS Fourier transformed infrared spectroscopy by diffuse reflectance
EBCT Empty bed contact time (min)
ENP Engineered nanoparticles
FeOOH Iron oxide nanoparticles
HRTEM High resolution transmission electron microscopy
K Langmuir isotherm constant (L/mg-min)
K1 Adsorption kinetic constant (L/mg-min)
K2 Desorption kinetic constant (L/min)
kL Liquid film mass transfer coefficient (cm/h)
k2 Constant rate of pseudo-second order sorption
(g/mg-min)
m Mass of adsorbent (mg)
MWNT Multiwall carbon nanotubes
nZVI Zero valent-iron nanoparticles
ox-GAC Nitric acid-oxidized activated carbon by 1 h ox-CNx Nitric acid-oxidized nitrogen-doped carbon
nanotubes
ox-CNx-1h Nitric acid-oxidized nitrogen-doped carbon nanotubes by 1h
ox-CNx-3h Nitric acid-oxidized itrogen-doped carbon nanotubes by 3h
ox-CNx-5h Nitric acid-oxidized nitrogen-doped carbon nanotubes by 5h
xxvi
PAC-F400 Powder activated carbon Filtrasorb 400
pHPZC pH of point of zero charge
pHSP Slurry pH
PZC Point of zero charge
Q Influent flow rate (cm3/min)
q Adsorption capacity (mg/g)
qe Adsorption capacity at equilibrium (mg/g)
qm Maximal surface concentration, Langmuir isotherm
constant (mg/g)
qmax Maximum adsorption capacity (mg/g)
q Mass of metal adsorbed per unit mass of
adsorbent (mg/g)
qm Langmuir isotherm constant (mg/g)
qt Adsorption capacity at time t (mg/g)
r Distance in radial direction of nanoparticle (cm)
R Radius of the nanoparticle (cm)
S Outer surface area of the adsorbent particle per
unit mass of adsorbent (cm2/g)
SBET Specific surface per unit mass of adsorbent
determined by BET method (m2/g)
SV Surface area of the adsorbent particle per unit
volume of the adsorbent particle (cm-1)
SEM Scanning electron microscopy
STEM Scanning transmission electron microscopy
SWNT Single wall carbon nanotubes Elicarb® from Thomas Swan
t Time (min)
T Absolute temperature (K)
TAS Total acidic sites (mmol/g)
TGA Thermo-gravimetric analysis
xxvii
Tm Melting temperature (°C)
V Volume of cadmium solution (L)
VA Molar volume of cadmium at its boiling temperature
(cm3/mol)
VP Pore volume per unit mass of nanoparticles
(cm3/g)
XRD X-ray diffraction
ρs Solid density of nanoparticles (g/cm3)
ρp Apparent density of nanoparticles (g/cm3)
p Void fraction of nanoparticle adsorbent (g/cm3)
τ Tortuosity factor
φ Dimensionless concentration of cadmium in
solution
φexp Experimental dimensionless concentration of
cadmium at equilibrium
φpred Dimensionless concentration of cadmium in
solution predicted with the diffusion model
φ∞ Dimensionless concentration of cadmium at
1
CHAPTER 1
Carbon Nanostructured Adsorbents for the Removal of Toxic
Metals from Aqueous Solution. State of the Art
1.1 INTRODUCTION
Water plays a key role in human societies. Proper management of water resources is an essential component of growth, social and economic development, poverty reduction and sustainable environmental development. The availability of water resources and their management is determinant in a country’s growth strategy (WWAP, 2009). As world population grows, there is a challenge of having sufficient water of the right quality, at the right place and at the right time. However, this
resource becomes increasingly scarce relative to demand. Great efforts are necessary to protect the water supply for humans, biodiversity and ecosystems (WWF4, 2006a).
That is why all countries are encouraged to optimize water consumption at the same time as preventing its pollution. Several actions are suggested, such as water recycling and reuse, adoption of international standards for water quality, and more investment in wastewater treatment and sanitation infrastructure, with preference for small-scale and local solutions. These actions require technological innovation for efficient operation of current water infrastructure, even more to design new cost-effective technologies (WWF4, 2006 b).
Advancement of knowledge and understanding of water treatment at different levels of sophistication has resulted in better technological options. Emerging nanotechnology can offer significant opportunities for the water sector. This new knowledge would result in using nanoparticles, nanofiltration or other products
2
reduce the complexity of large centralized systems (Savage and Diallo, 2005; WWAP, 2006; 2009).
1.1.1 Water Pollution by Toxic Metals and Technologies for Treatment
Population growth has increased water demand for almost all productive activities such as agriculture, industry, energy and transportation. The wastewater from these activities containing high loads of toxic substances such as toxic metals is frequently discharged into water bodies. These contaminants have the potential for long-range transport, dispersal and deposition. Nowadays most bodies of water, either surface or groundwater sources, are polluted with toxic metals, which persist for long periods even at low concentrations (WWAP, 2003).
Toxic metals cannot be destroyed or broken down through treatment or environmental degradation, leading to different human health problems such as lead poisoning and cancer (USEPA, 1999). International standards for water quality list as priority toxic metals: lead, cadmium, mercury, chromium, nickel, silver, thallium and zinc. These standards are guideline values that are reasonably achievable through practical treatment technologies. However, these values are revised continuously. It is necessary to improve current technologies and to develop new ones (WHO, 2006).
In this research, lead and cadmium were selected as model toxic metal pollutants of water. A brief description of these metals, water pollution sources, and toxicology are presented in the following sections.
1.1.1.1 Water Pollution by Lead
3
Lead is principally used in the production of lead-acid batteries, solder and alloys. Lead is rarely present in tap water as a result of its dissolution from natural sources; rather, its presence is primarily due to plumbing systems containing lead in pipes, solder, fittings or the service connections. The amount of lead dissolved from the plumbing system depends on several factors including pH, temperature and hardness of the water. Lead concentrations in drinking-water are generally below 5 mg/L, although much higher concentrations (above 100 mg/L) have been measured where lead fittings are present. Infants, children up to 6 years of age and pregnant women are most susceptible to its adverse health effects. Lead accumulates in the skeleton impeding calcium metabolism, both directly and by interfering with vitamin D metabolism. Lead is toxic to the central and peripheral nervous systems, inducing subencephalopathic neurological and behavioral effects. Lead and its compounds are classified as possible human carcinogen. However, there is evidence from studies in humans that adverse neurotoxic effects other than cancer may occur at very low concentrations of lead and that a guideline value derived on this basis would also be protective for carcinogenic effects. Most lead in drinking-water arises from plumbing, and the remedy consists principally of removing plumbing and fittings containing lead (WHO, 2006).
Lead in aqueous solutions participates in a series of consecutive proton transfers that determine their chemical compounds in solution. The lead speciation diagram (Fig. 1.1) shows the distribution of species present in aqueous solution as a function of pH. Lead ion (Pb2+) is stable below pH 5.5, then species PbOH+, Pb(OH)2, Pb(OH)3- are consecutively formed as a result of hydrolysis reactions
4 Fig. 1.1 Species distribution of 4x10-4 M Pb (II) in aqueous solution at 25 °C.
1.1.1.2 Water Pollution by Cadmium
The World Health Organization recommends a threshold limit guideline value of 0.003 mg/L for cadmium in drinking water (WHO, 2006). This value can vary in each country in function of available technologies. In Mexico, the corresponding value is 0.005 mg/L (NOM-127-SSA1-1994, 2000).
Cadmium metal and its compounds are used in the steel industry, plastics and in batteries. Cadmium is released to the environment in wastewater, and diffuse pollution is caused by contamination from fertilizers and local air pollution. Contamination of drinking-water may also be caused by impurities in the zinc of galvanized pipes and solders, and some metal fittings.
5
Cadmium accumulates primarily in the kidneys and has a long biological half-life in humans of 10 to 35 years. There is evidence that cadmium is carcinogenic when it is inhaled. However, there is no evidence of carcinogenicity by ingestion and no clear evidence for the genotoxicity of cadmium. The kidney is the main target organ for cadmium toxicity. The critical cadmium concentration in the renal cortex is about 200 mg/kg and would be reached after a daily dietary intake of about 175 μg
per person for 50 years.
Speciation of cadmium in aqueous solution (Fig. 1.2) shows that at a concentration of 7x10-4 M (78 mg/L), cadmium ion (Cd2+) is stable below pH 7.8; then species CdOH+, Cd(OH)2, Pb(OH)3- are consecutively formed as result of
hydrolysis reactions (Stumm and Morgan, 1996).
6 1.1.1.3 Water Treatment Technologies Available to Remove Toxic Metals
Drinking water purification and wastewater treatment will invariably be required to reduce both health risks and negative environmental impacts. Different technologies can be employed to remove pollutants as toxic metals: chemical precipitation, coagulation-flocculation, membrane filtration, ion exchange and adsorption. A brief review of advantages and limitations of these technologies are described as follows (Kurniawan et al., 2006):
a) Chemical precipitation. After pH adjustment to the basic conditions (pH 11),
the dissolved metal ions are converted to the insoluble solid phase via a chemical reaction with a precipitant agent such as lime, lime plus sodium carbonate, or sodium hydroxide. Typically the precipitated metal from the solution is in the form of hydroxides, sulfides, or carbonates. The precipitate can then be removed from the treated water by clarification (settling) and/or filtration. Lime precipitation can be employed to effectively treat inorganic effluent with a metal concentration higher than 1000 mg/L. Other advantages are the simplicity of the method, inexpensive equipment requirement and wide availability of chemicals used. The main disadvantages are the necessity for large amounts of chemicals to reduce metals to an acceptable level for discharge. Since precipitated metals cannot be economically recovered from sludge, this requires further treatment increasing cost due to sludge disposal. Other disadvantages are slow metal precipitation, poor settling and aggregation of metal precipitates, and the long-term environmental impacts of sludge disposal.
b) Coagulation-flocculation. In coagulation colloidal particles are destabilized by
7
neutralization, adsorption and entrapment. The floc is removed from the treated water by subsequent processes such as sedimentation, flotation or filtration. This method can treat effluents with a metal concentration between 100 to 1000 mg/L, at pH ranging from 11 to 11.5 for optimum efficiency. The main advantages of the method are improved sludge settling, bacterial inactivation capability and sludge stability. The disadvantages are a high operational cost due to chemical consumption and further requirement for final disposal of sludge.
c) Membrane processes. Membranes have become an increasingly available filtration technology which can separate a wide range of substances organic and inorganic present in water. They can also be used in industrial and drinking water treatment, wastewater treatment and brackish and seawater desalination, at costs that are being reduced substantially. In function of the size of the particle to be retained, various types of membrane filtration can be employed: ultrafiltration (UF), nanofiltration (NF) and reverse osmosis (RO).
8
pH range from 3 to 11 and 5 to 15 bars of pressure. Pressure is the major parameter controlling effectiveness of metal removal: a higher pressure ensures higher metal removal, and thus higher energy consumption. Other advantages of RO include a high water reflux rate, high salt rejection, resistance to biological attack, mechanical strength, chemical stability and the ability to withstand high temperatures.
The main disadvantages of membrane processes are the decrease in performance due to membrane fouling, with adverse effects on the membrane system causing flux decline, an increase in transmembrane pressure and the biodegradation of the membrane materials, resulting in a high operational of cost for the system.
d) Ion exchange. In this process a reversible exchange of ions between the solid
and liquid phases occurs, where an insoluble substance (resin) removes ions from a solution, and releases other ions of like charge in a chemically equivalent amount without any structural change of the resin. The process takes place without the formation of chemical bonds, hence it is possible to recover the metal loaded by using adequate reagents (Zagorodni, 2007). Ion exchange resins are cross-linked polymers carrying fixed functional groups or sites, usually sulfonic acid (-RSO3ˉ).
Ion exchange is effective to treat inorganic effluent with a metal concentration in the range from 10 to 100 mg/L, in acidic conditions at pH ranging from 2 to 6. This process does not present sludge disposal problems. Appropriate pretreatment of the effluent is required prior to using an ion exchange process, as the removal of suspended solids.
9 e) Adsorption. This is a mass transfer operation by which a substance is
transferred from the liquid phase to the surface of a solid and becomes bound by physical and/ or chemical interactions. Adsorption is used in drinking water treatment as tertiary process, for the removal of color and odor for achieving regulatory requirements. The primary adsorbent materials used in the adsorption process for drinking water treatment are activated carbons (Suzuki, 1990; Bansal et al., 1998; Bansal and Goyal, 2005). Recently research has pointed out low-cost adsorbents derived from agricultural waste, industrial by-products or natural materials for the removal of heavy metals. Use of granular activated carbon (GAC) is considered to be the best available technology for removing low-solubility contaminants such as toxic metals in trace concentrations (Radovic, 2001). A broader description of this technology is included in the following section.
1.1.1.4 Adsorbents for Toxic Metals
Some adsorbent materials are capable to remove toxic metal from aqueous solutions such as activated carbons, zeolites, clays and polymeric resins.
a) Zeolites are porous crystalline aluminosilicates which comprise assemblies of SiO4 and AIO4 tetrahedra joined together through the sharing of oxygen atoms.
Commercially the most important are chabazite, faujasite and mordenite. Cavities contained within the zeolite framework are connected by regular channels of molecular dimensions, into which adsorbate molecules can penetrate. The internal porosity is high and thus the majority of adsorption takes place internally. For this reason zeolites are capable of effectively remove adsorbates on the basis of size, shape and other properties such as polarity.
10
Fuller's earth is an activated natural montmorillonite. Its pore size is altered and its surface area increased (150-250 m2/g) by acid treatment. It is relatively inexpensive and which it is in cationic form is capable of adsorbing a range of polar molecules as toxic metal ions (Thomas and Crittenden, 1998).
c) Polymeric resins are cross-linked polymers carrying fixed ions which can be stoichiometrically exchanged for other ions of the same sign. Some groups are able to form chelate structures with metal ions. The most typical example of ion exchange resin is sulphonated polystyrene cross-linked with divinylbenzene. The group –SO3H can exchange the hydrogen ion for any other cation, aditionally it can
bear a wide diversity of functional groups (Zagorodni, 2007).
1.1.1.5 Adsorption of Toxic Metals on Activated Carbon
Activated carbon (AC) has been proven to be an effective adsorbent for the removal of a wide variety of organic compounds and heavy metals dissolved in aqueous media (Yin et al., 2007). Activated carbon is produced by the combustion of carbonaceous material, normally wood, coal, coconut shells or peat. This activation produces a porous material with a large surface area (500–1500 m2/g)
(Bansal et al., 1998; Bansal and Goyal, 2005).
11
adsorption of heavy metals due to reactions to form complexes as indicated by the equation 1.1 (Yin et al, 2007):
(1.1)
Fig. 1.3 Representation of oxygen functional groups present on the carbon
activated surface.
12
the amount of surface oxygen-containing groups (Contescu et al., 1997; Radovic, 2001; Bansal and Goyal, 2005).
Adsorption of heavy metals in solution depends very much on pH and it has been found that it occurs at 1 to 2 pH units below the value that is required for precipitation of the metal hydroxide. In general, activated carbons are capable of adsorbing many of the toxic metals such as copper, lead, zinc, nickel, cadmium, mercury, chromium, cobalt and silver. Many works have been reviewed and discussed by several authors (Radovic, 2001; Bansal and Goyal, 2005).
Adsorption at the solution-carbon interface is already widely used on large scale water treatment to remove toxic metals. Carbon capacity is strongly dependent on the water source and is greatly reduced by the presence of background organic compounds (Cooney, 1999). Activated carbons are normally used either in powdered (PAC) or in granular (GAC) form. The choice between PAC and GAC will depend upon the frequency and dosage required. PAC would be preferred where low dosage rates are required. PAC is dosed as slurry into the water and is removed by subsequent treatment processes. Its use is therefore restricted to surface water treatment with existing filters. GAC in fixed-bed adsorbers is used much more efficiently than PAC dosed into the water. The service life of a GAC bed is dependent on the capacity of the carbon, and the contact time between the water and the carbon (empty bed contact time, EBCT). EBCT is controlled by the flow rate of the water, often values for EBCTs are in the range from 5 to 30 min. Metal ions exhausted activated carbons are normally used only once before their disposal (WHO, 2006).
13
1.2 NANOTECHNOLOGY DEVELOPMENT AND POTENTIAL
APPLICATIONS TO WATER TREATMENT
The expansion of scientific knowledge and technological applications is changing the way in which water is used, cleaned and reused, to meet human, economic and environmental needs. Nanotechnology has been emerging rapidly; it has been predicted that it will exert strong influence on the technologies related to supply, use and management of water resources (WWAP, 2009).
Nanotechnology is the manipulation of matter for particular applications at the atomic, molecular, or macromolecular levels, using a length scale of approximately one to one hundred nanometers in any dimension (USEPA, 2007). Nanotechnology can develop new materials with enhanced properties and attributes, resulting of larger surface area per unit of volume and quantum effects that occur at the nanometer scale (Renn and Roco, 2006).
Applications of nanotechnology will penetrate nearly all sectors and spheres of life (communication, health, labour, mobility, housing, relaxation, energy, food) and will be accompanied by changes in the social, economic, ethical and ecological spheres. Already nanomaterial-containing products are available in U.S. markets including coatings, computers, clothing, cosmetics, sports equipment and medical devices (Renn and Roco, 2006; Hristozov and Malsch, 2009).
Nanomaterials are all materials with sizes on the nanoscale in at least one of their dimensions, while nanoparticles (NP) are nanosized in at least two dimensions. The nomenclature nanoparticle includes particles as well as fibrous materials and tubes, but it excludes materials, such as coatings, films and multilayers. Two types of nanoparticles can be distinguished: naturally occurring ones such as those produced in volcanoes, forest fires or as combustion by-products, and engineered nanoparticles (ENP), deliberately developed to be used in application (Hristozov and Malsch, 2009).
14
a) Carbon-based materials: Composed mostly of carbon; spherical and ellipsoidal carbon nanomaterials are referred to as fullerenes, while cylindrical ones are called nanotubes.
b) Metal-based materials: Include quantum dots, nanogold, nanosilver and metal oxides.
c) Dendrimers: Are nanosized polymers built from branched units. The surface of a dendrimer has numerous chain ends, which can be tailored to perform specific chemical functions.
d) Composites: Combine nanoparticles with other nanoparticles or with larger, bulk-type materials.
Characteristics of nanomaterials such as enhanced reactivity, surface area, subsurface transport, and/or sequestration, could benefit environmental remediation by faster and more economical processes than current conventional approaches. It is necessary to understand how to best apply nanotechnology for pollution prevention as well as in environmental detection, monitoring and remediation. Environmental remediation is the degradation or sequestration of chemical and radiological contaminant, reducing the risks to human and environmental receptors (USEPA, 2007).
Nanotechnology shows particular promise for water resources, to improve water quality through water treatment or remediation. Key areas are desalinization, water purification, wastewater treatment and monitoring. The first three areas involve the use of nanofiltration technology, nanomaterials and nanoparticles to remove or reduce water contaminants. Monitoring involves the use of new sensor technology coupled with micro and nanofabrication technology. Nanosensors would be highly accurate and portable for detection of biological and chemical water contaminants, present in very low concentrations (Tour, 2007; WWAP, 2009).
15
use of heavily polluted and saline water for drinking, sanitation and irrigation (Weber, 2002; Savage and Diallo, 2005). ENP applied to water treatment might control the destination, transport, and bioavailability of toxic metals and other contaminants through adsorption, redox reaction, and other biogeochemical processes. However, the stability and aggregation behavior of ENP in water could impact their reactivity and efficacy in contaminant treatment (He et al., 2008).
There are several reviews about using ENP and nanomaterials to remove metal ions from aqueous solutions or wastewater. So far the most promising results have been obtained using carbonaceous nanomaterials (fullerenes, SWNT and/ or MWNT), biopolymers, zero-valent iron and nanocrystalline titanium oxide (TiO2)
(Savage and Diallo, 2005; Li et al., 2007; Mauter and Elimelech, 2008; Shannon et al., 2008; Theron et al., 2008).
Applications involving dispersive uses of nanomaterials in water have the potential for wide exposures to aquatic life and humans (USEPA, 2007). It is possible that ENP can cause novel environmental problems and/or impose risks to human health. It is very important to study the environmental destination of ENP in order to understand their pathways of environmental and human exposure. At this point of time and stage of knowledge, it is impossible to make any collective judgment about the potential risks of exposure to nanomaterials (Renn and Roco, 2006).
The destination of ENP in water is determined by factors such as aqueous solubility, reactivity of the ENP with the chemical environment and their interaction with certain biological processes. Some ENP might be removed from water by processes such as sorption to soil and sediment particles, biotic and/ or abiotic degradation (hydrolysis and photocatalysis). Nonetheless, some insoluble ENP can be stabilized in aquatic environments by interaction with natural organic matter (NOM) or humic acids (Hristozov and Malsch, 2009).
16
be more accurate. For all types of ENPs, the most suitable dose-descriptors need to be determined in terms of their surface area, mass, morphology and chemical composition (Renn and Roco, 2006; Hristozov and Malsch, 2009). Considering this situation, adequate immobilization of nanoparticles before their use in water-related applications could prevent their dispersion, and their probable adverse environmental effects.
Nanotechnology-based applications to the water sector are still under fundamental research, and it is unclear when they will be ready for wide-scale use. These new technologies must compete against currently used technologies, looking for cheaper solutions that meet government standards. In addition, it is necessary to overcome factors such as availability of nanomaterials in large scale and lack of knowledge and regulations about their environmental and health impacts (WWAP, 2009).
1.2.1 Adsorbent Nanoparticles for Toxic Metals
Research and development of novel materials with increased capacity, affinity and selectivity for heavy metals and other water pollutants is being performed. Benefits from using nanomaterials may derive from their enhanced reactivity, surface area and sequestration characteristics. Some materials possessing unique functionalities potentially applicable to heavy metal remediation and purification from water are carbon nanotubes, biopolymers and zero-valent iron nanoparticles.
17
of metals, the aggregated polymers can be reused for subsequent cycles (Kostal et al., 2005).
Iron nanoparticle technology represents one of the first generation nanoscale environmental technologies, receiving considerable attention for its potential applications in groundwater treatment and site remediation. The greatest interest is in the use of zero-valent iron nanoparticles (Fe0; nZVI), for which reactivity is driven by oxidation of the Fe0 core. Their higher reactivity compared to their microscale counterparts has been attributed to the greater density and higher intrinsic reactivity of their surface sites. Extensive laboratory studies have demonstrated that nZVI nanoparticles are effective for the transformation heavy metal ions such as lead, copper, nickel, and chromium. Both, reduction and surface complex formation are reported as the mechanisms for ion transformation (Li and Zhang, 2006).
1.2.2 Carbon Nanotubes: Synthesis, Properties and Modification
Carbon nanotubes (CNT) are key elements in nanotechnology. The structure of carbon nanotubes is depicted as a rolled segment of a single graphite layer, called a graphene sheet (Fig. 1.4). It is formed by linking each carbon atom to three equivalent neighbors in a trigonal planar fashion (C-C distance, 1.42 Å) (Dresselhaus, 1997; Dresselhaus et al., 2001). Carbon nanotubes structured by
18 Fig. 1.4 Fundamental structure for (A) graphene sheet, (B) single-wall carbon
nanotube, SWNT, (C) multiwall carbon nanotube, MWNT (modified figure from Mauter and Elimelech, 2008) and (D) typical defects and functional groups in an oxidized carbon nanotube (modified figure from Hirsch, 2002).
Synthesis of carbon nanotubes can be carried out by several techniques, at high temperatures (above 3200 °C) or at medium temperatures (lower than 1000 °C). High-temperature routes involve sublimating graphite in a reduced atmosphere of rare (inert) gases, and condensing resulting vapor under a high temperature gradient. Used methods for sublimating graphite could be: an electric arc formed between two electrodes made in graphite, an ablation induced by a pulsed laser, or a vaporization induced by a solar or a continuous laser beam. Medium-temperature routes are based on catalytic chemical vapor deposition (CVD) processes. CVD make possible the growth of carbon nanotubes of various sizes and shapes, from carbon-containing gaseous compounds which decompose catalytically on transition-metal particles (Loiseau et al., 2006).
19
produced with metallic catalyst whichever the synthesis process used (Dresselhaus et al., 2001).
Carbon nanotubes have a very stable structure due to sp2 covalent bonds between carbon atoms. Moreover, cylindrical curvature gives some sp3 character to the C–C bonding, resulting in the richness and diversity of the physical and chemical properties of carbon nanotubes. Only perfect graphene sheets are chemically inert, but surface defects existing on carbon nanotubes constitute reactive sites (Niyogi et al., 2002; Tománek et al., 2008). It is possible to modify and control the physicochemical properties of carbon nanotubes by doping processes, which consist of introducing either non-carbon atoms or molecules at small concentrations, in the plane of the graphene network. This process can tailor the electronic, vibrational, chemical and/ or mechanical properties of nanotubes, useful for a wide range of applications (Terrones et al., 2002; 2008).
20
Besides doping, chemical modification of carbon nanotubes with different functional groups for specific applications is a major growth area nowadays (Tománek et al., 2008; Wang et al., 2009). Functionalization can be used to modify the interface between the environment and the outer wall of CNTs (Fig. 1.4D). This would, for example, modify solubility and facilitate dispersion in a given medium. The tips are the most reactive part of the CNTs because it is where the highest strains are located and where the carbon atoms have some sp3character (Hirsch, 2002). All oxidizing treatments involve the use of acidic solutions or thermal treatment (HNO3, KMnO4 + H2SO4, K2Cr2O7, H2O2, CO2, O2), which will attack the
tips in order of priority, but structural defects located on the walls (heptagon-pentagon pairs for example) will be attacked as well to a smaller extent (Niyogi et al., 2002; Ovejero et al., 2006). All these treatments will degrade the carbon nanotubes depending on the strength of the oxidizing agent and the duration of the treatment. Oxidation of CNTs leads to the formation of chemical groups such as carboxyl functions, lactones, ketones, or hydroxyl groups. These organic functions can be used to graft other functional groups or molecules (Hirsch, 2002; Tasis et al., 2006).
1.2.3 Adsorption of Toxic Metals on Carbon Nanotubes
21
their maximum adsorption capacity was 27.6 mg/g at pH 5.2 (Li et al., 2003b). Also different reagents have been used to modify the nanotubes: hydrogen peroxide (H2O2), potassium permanganate (KMnO4) and nitric acid (HNO3). The oxidized
nanotubes removed cadmium with maximum adsorption capacities of 2.6, 5.1 and 11 mg/g, respectively, at the cadmium equilibrium concentration of 4 mg/L (Li et al., 2003 c). Nitric acid solutions are the most used oxidizing agent to attach oxygen-containing groups onto carbon nanotubes to modify their surface properties for metal ions adsorption. Furthermore, adsorption equilibrium tests at different temperatures were conducted to evaluate the effect on adsorption capacity; it was found that adsorption with carbon nanotubes is an endothermic process (Li et al., 2005). Also, carbon nanotubes synthesized by using different conditions of precursor, catalyst and temperature (xylene-Fe-800°C, benzene-Fe-1150°C, propylene-Ni-750°C or methane-Ni-650°C) were tested to adsorb lead. It was found that those synthesized with methane-Ni-650°C had a higher lead adsorption capacity (82.6 mg/g at pH 5), because their less crystalline structure and higher concentration of carboxylic groups (Li et al., 2006).
The adsorption capacity of MWNT for several metal ions has been evaluated at different conditions of ionic strength, pH, temperature, foreign ions, etc. Some of these studies include kinetics analysis of the experimental data by the pseudo-second order reaction model. Published results for maximum nickel adsorption capacity are 9.8 mg/g at 30 °C and pH 6.5 (Chen and Wang, 2006), 49.6 mg/g at room temperature and pH 6 (Kandah and Meunier, 2007), 38.46 mg/g at room temperature and pH 7, for MWNT oxidized by sodium hypochlorite (Lu et al., 2008), and 7.4 mg/g due to the effect of polyacrylic acid (PAA), with a structure like that of natural organic matter (NOM) (Yang et al., 2009).
22
and pH 7 (Lu et al., 2006). The maximum nickel adsorption capacity onto oxidized SWNT is 47.85 mg/g at room temperature and pH 7 versus 38.46 mg/g for oxidized MWNT (Lu et al., 2008).
Recently, two adsorption studies were conducted with nitrogen-doped carbon nanotubes (CNx) to remove metal ions from aqueous solutions. Maximum adsorption capacity of 31 mg/g (Andrade-Espinosa et al., 2008) and 21.6 mg/g, at pH 7 (Perez -Aguilar et al., 2010) were obtained for cadmium and 28.9 mg/g at pH 5 for lead (Perez -Aguilar et al., 2010).
In general, carbon nanotubes have shown better adsorption capacities for toxic metal ions in aqueous solution than activated carbons. Their adsorptive features besides a more efficient regeneration, and shorter time to reach equilibrium than activated carbon, have led to a search for an adequate use of CNT in devices as membranes and filters for water purification systems (Rao et al., 2007; Stafiej and Pyrzynska, 2007; Wang et al., 2007a,b; Yang et al., 2009).
23
Furthermore, considering the risk of liberating carbon nanotubes to water sources, it is necessary to effectively immobilize these nanostructures, in a way that their application is possible in adsorber or filter devices for water treatment.
1.3 MOTIVATION FOR THIS RESEARCH
Nanotechnology is evolving rapidly, offering the possibility to design and obtain customized materials for specific applications from a molecular level, more accurately and economic than current bulk methods.
The development of highly efficient nanotechnology-based water treatment and purification systems will occur in the near future. These will offer higher reliability of treatment, enabling reduction of equipment size, more energy saving, simplification and cost reduction compared to conventional technologies.
Carbon nanotubes are key nanomaterials in nanotechnology. The graphene sheet is the fundamental structure of carbon nanotubes, likewise in activated carbon, the most used adsorbent. Hence, the development of carbon nanotube adsorbents for water applications is of special interest. Current knowledge on carbon chemistry has allowed the modification of graphene sheets, and consequently the chemical properties of carbon nanotubes. Hence, there is a broad range of possibilities to design carbon nanotubes for specific uses and/or to remove certain pollutants. A better performance of the carbon nanotube as adsorbents is expected compared to currently used activated carbons.
24
Finally, there are challenges and difficulties associated with the dissemination of nanotechnology, which is especially important for developing countries such as Mexico. Most of the time new technologies are very expensive because they are imported from industrialized countries. In addition, scarcity of human resources with appropriate technical knowledge and abilities might prevent the optimum operative efficiency of these novel tools. These are the main reasons that motivated exploring and contributing to the current knowledge of using nanotechnology in water treatment.
1.4 GENERAL AND SPECIFIC OBJECTIVES
This research is focused on nitrogen doped and nitric acid oxidized carbon nanotubes. The general aim is to evaluate the adsorption performance of carbon nanotubes to remove toxic metals present in aqueous solutions. This objective was covered from two different approaches: through batch adsorption experiments by using suspended nanotubes, and dynamic adsorption experiments by using bed packed columns with supported nanotubes in a polymeric composite.
To achieve the main goal, the following specific objectives were established: selecting cadmium and lead in aqueous solution as model metal ions.
I. To conduct the physical and chemical characterization of carbon nanotubes by several techniques to understand interactions between the nanotubes surface and heavy metal ions in aqueous solution, in order to suggest an adsorption mechanism.
II. To evaluate the adsorption capacity at equilibrium of modified novel nitrogen-doped carbon nanotubes (CNx), and to compare their performance with single carbon nanotubes (SWNT), multiwall carbon nanotubes (MWNT), as well as with commercially available activated carbon.